smithery.ai

translate-vtt

Translate an English VTT subtitle file to Korean

First seen Mar 25, 2026

Installation

$ npx skills add https://smithery.ai

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery.ai · top by installs.

npx skills add https://smithery.ai

Browse all from smithery.ai

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 2,586 B
  • docs SUMMARY.md 69 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

VTT Subtitle Translation

Translate the given English VTT subtitle file to Korean using translate-vtt.py.

Step 1: Extract

Run:

python3 translate-vtt.py extract "$ARGUMENTS"

Note the batch count and total unique texts from the output.

Step 2: Translate (Parallel)

Launch one Task agent (subagent_type: general-purpose) per batch file. All tasks MUST be launched in a single message (parallel execution).

For each batchN.json (N = 0 to batchcount - 1), use this prompt:

Read batch_N.json in the subtitles/ directory (same directory as the VTT file).
It contains English subtitle lines that need Korean translation.

For each entry in the "entries" array:
1. Keep all existing fields (id, text) exactly as-is
2. Add a "translation" field with the Korean translation

Write the result to trans_N.json in the same subtitles/ directory.
Keep all metadata fields (source_file, batch_index, total_batches) unchanged.

Translation rules:
- Natural conversational Korean (YouTube tutorial tone)
- Keep technical terms in English: API, GitHub, CLI, MCP, JSON, npm, git,
  TypeScript, JavaScript, React, Node.js, VS Code, Docker, SDK, etc.
- Keep proper nouns in English (people names, product names, company names)
- Numbers and units stay as-is
- Each translation is a single line (no line breaks)
- Translate EVERY entry without exception
- Output valid JSON with Korean characters (not unicode escapes)

Wait for ALL agents to complete before proceeding.

Step 3: Reconstruct

python3 translate-vtt.py reconstruct "$ARGUMENTS"

Step 4: Verify

Check:

  • Output .ko.vtt file exists
  • Cue count matches original
  • Language header is ko
  • Spot-check first and last few cues

Step 5: Comparison HTML

Generate a self-contained, searchable side-by-side EN/KO comparison:

python3 translate-vtt.py compare "$ARGUMENTS"

Output: <name>.ko.compare.html next to the VTT files. Report the path to the user so they can review the translation quality in a browser.

Step 6: Cleanup

python3 translate-vtt.py cleanup subtitles

Cleanup only removes intermediate JSON files; the .ko.vtt and .compare.html deliverables are kept.

Error Recovery

  • If a Task agent fails, re-run only that specific batch
  • If reconstruct reports missing translations, check which trans_N.json is incomplete
  • The extract step is idempotent (safe to re-run)